The Science of Batting Order - Is the Best Hitter in the Third Spot Really Optimal?

Japanese Baseball's Batting Order Convention

NPB batting order conventions have been rigid for decades. The leadoff hitter is a speedy on-base type, the second hitter sacrifices runners over, the third is the team's best hitter, cleanup provides power, and fifth drives in remaining runners. This hierarchy aims to score through the three-four-five heart of the order, with the top two reaching base for the middle to drive in. The third spot demands versatility in all situations, and NPB's greatest hitters including Hiromitsu Ochiai, Ichiro, and Hideki Matsui made their names batting third.

The Second-Spot Revolution Theory

Sabermetric research challenges conventional wisdom. Simulations show that placing the highest-ability hitters in the first and second spots, which receive the most plate appearances, maximizes season-long run production. The best-hitter-second theory gained MLB traction in the late 2010s. The logic is clear: the second hitter bats immediately after the leadoff hitter reaches base, increasing runner-on-base opportunities, and receives approximately 18 more plate appearances per season than the third hitter. MLB stars like Mike Trout and Mookie Betts now routinely bat second.

NPB's Evolving Second Spot

NPB's second-hitter role is gradually shifting. The sacrifice-bunt convention is eroding, with more managers placing offensive threats second. Second-hitter sacrifice totals have dropped significantly from a decade ago while on-base and slugging percentages have risen. However, placing the best hitter second remains rare in NPB. Many managers were educated in the sacrifice-second era and resist philosophical change. Japanese baseball culture also treats batting third or fourth as status symbols, creating risk that moving a star to second is perceived as demotion.

The Sacrifice Bunt Cost-Benefit Analysis

Sacrifice bunt analysis is inseparable from the second-hitter debate. Statistically, bunting with no outs and a runner on first reduces run expectancy from approximately 0.85 to 0.70 runs. The bunt marginally increases the probability of scoring exactly one run while substantially reducing multi-run potential. NPB still employs sacrifices at two to three times MLB's rate. Late-inning sacrifices in close games have tactical justification, but early-inning bunts frequently reduce scoring efficiency. Data-driven lineup optimization and sacrifice reduction are two sides of the same coin.

Does an Optimal Batting Order Exist?

Optimization research has advanced but no single correct answer exists. Simulation-derived optimal lineups maximize average expected runs, but individual games involve pitcher matchups, handedness combinations, and park factors. Batting position also affects player psychology: some thrive under the prestige of batting third or fourth while others wilt under pressure. Data-optimal and real-world-optimal lineups do not always align. NPB lineups will continue evolving, but fully data-driven construction will take time. The key is questioning the best-hitter-third assumption and building lineups flexibly based on roster composition.

Fixed Lineups versus Daily Rotation

NPB managers employ two philosophies in lineup construction: fixed and rotating. The fixed approach keeps the batting order virtually unchanged throughout the season, imprinting each player's role. Yakult's Katsuya Nomura in the 1990s exemplified this style, anchoring Atsuya Furuta at cleanup to create rhythm and stability. The rotating approach reshuffles the order each game based on the opposing pitcher's handedness and pitch mix. Hiroshima Carp manager Koichi Ogata used over one hundred different lineups during the 2018 season. Statistically, fixed lineups produce lower variance in seasonal results and greater stability, while rotating lineups can disrupt opponents' preparations in short playoff series. Regardless of which method a manager chooses, unexamined rigidity is intellectual laziness. Intentional, evidence-based selection is the essence of batting-order science.

The Pitcher's At-Bat and the DH Rule's Impact

No discussion of batting-order science is complete without addressing the designated hitter rule. Through 2024, the Central League maintained the system where pitchers batted ninth, imposing a major constraint on lineup construction. With a pitcher offering negligible offensive production in the ninth spot, the eighth hitter effectively became a secondary bottom-of-order slot, severing connectivity to the leadoff hitter. After the Pacific League adopted the DH in 1975, structural differences in lineup philosophy emerged between the two leagues. In the DH-equipped Pacific League, all nine spots demand hitting ability, making a seamless lineup from first through ninth the ideal. In the Central League, the pitcher's at-bat was accepted as a wasted plate appearance, and how managers constructed the surrounding order showcased their craft. With the Central League adopting the DH from 2025, lineup-construction freedom has expanded across all of NPB. This institutional change provides an opportunity to fundamentally restructure traditional batting-order conventions.

Lineup Construction in the Data Era and Player Psychology

With tracking data and Statcast proliferating, hitter characteristics are quantified in unprecedented detail. Combining metrics such as exit velocity, launch angle, swing path, and zone-specific batting averages theoretically allows precise calculation of batter compatibility and expected runs by lineup position. Yet baseball is played by humans, and psychological factors cannot be eliminated. Batting position directly influences motivation and focus. When the Orix Buffaloes won the Japan Series in 2022, manager Satoshi Nakajima kept Masataka Yoshida fixed at third despite analyses suggesting second was optimal given his on-base rate. Nakajima judged that Yoshida's self-awareness as a third-place hitter unlocked his full potential. The ultimate optimal lineup lies at the intersection of mathematical models and player psychology. When data-driven expected value aligns with the position where a player performs best, the lineup achieves maximum destructive power.